1

Finite Element Study of the Mass Transfer in Annular Reactor

Yehia M. S. El Shazly*1, and Sarah W. Eletriby
1
Alexandria University, Faculty of Engineering, Chemical Engineering Department, Egypt.
*yehia.elshazly@alexu.edu.eg; yehia.elshazly@hotmail.com
Abstract: The annular reactor is a very useful
design to carry many chemical reactions. Its
advantages such the low pressure drop and ease
of temperature control renders it more desirable
than traditional reactor, for example in the use as
a double heat exchanger, a dialyzers or a photocatalytic reactors, where a UV lamp can be
placed in the core of the reactor. In this study,
the isothermal mass transfer from the inner side
of the outer tube of the annular reactor is being
studied by the Finite Element technique within a
range of flow rates corresponding to
200<Re<12000. The study focuses on the effect
of the geometry on the rate of mass transfer in
the developing flow and in the fully developed
zones. The results showed that the surface mass
transfer is highest at the entrance and decreases
as the flow becomes fully developed. Also, the
annulus diameter ratio and the inlet section
dimensions were found to have an important
impact on the efficiency of the reactor. These
results are in agreement with previous CFD and
experimental studies.
Keywords: Annular Reactor, Solid Liquid Mass
Transfer, Finite Element, Developing flow, Fully
developed flow.

1. Introduction
Annular reactors are being increasingly in used
in chemical processes [1-4]. Its design offers
many recognized advantages such as low
pressure drop, large surface area to volume, easy
temperature control, and nearly isothermal
reaction conditions. It is important that when
designing these reactors to consider the effect of
the geometry on the flow pattern and its impact
on mass and heat transfer occurring in the
reactor. The study of the mass transfer has many
practical applications: for example for the case of
catalytic reactions where the rate of reaction is
controlled
by
the
diffusion
of
the
reactants/products to or away from the surface of
the catalyst. Also for the diffusion controlled
corrosion where the rate of corrosion is
controlled by the diffusion of the oxidizer to the
surface[5]. Moreover, for the case of annular
heat exchanger, the analogy between heat and

mass transfer enables us to deduce heat or mass

transfer from the other [6, 7].
Usually the rate of mass transfer is reported in
the form of dimensional relationships relating
Sherwood Number (the ratio of convective to
diffusive mass transfer) with Reynolds (ratio of
inertial force to viscous force) and Schmidt
Numbers (the ratio of momentum diffusivity to
molecular diffusivity). These correlation have
been obtained experimentally by many
techniques: the chemical reaction between the
liquid and solid [8-10], the dissolution of the
sparingly soluble solid into liquid[11-13], the
electrochemical technique [14-17], adsorption
[18, 19], and even have been followed with the
scanning electron microscopy [20]. However,
these correlations have the drawback that they
are applicable only to a specific reactor
configuration operated under a certain range of
hydrodynamic conditions[21, 22].
CFD has also been used [23-26]: if well
formulated, CFD has the advantages of being
time and cost consuming, coupled to being nonintrusive and able to predict the results of tests
that are not
easily performed. Several
mathematical models have been proposed and
evaluated which included mass transfer in
annular reactors used for different applications,
but their systems were limited either to laminar,
or to fully developed turbulent flow
conditions[27-30]. However, due to the design of
annular reactor, a fully developed flow is
unattainable. Fortunately, developing flow is
desirable as it yields favorable mass transfer
conditions. On the other hand, mass transfer
modelling under developing flow condition is a
complex task. Esteban et al [31]studied the
difference between the U and L entrance shape
for the annular reactor on the rate of mass
transfer using the finite volume technique and
concluded that though most of the mass transfer
took place near the entrance region of the
reactor, but both of the two shape gave similar
efficiency. They also found that among the
different turbulence models, the Low Reynolds
Number model gave the closest results to the
experiments. In this study, the commercial CFD
Comsol was used to study the effect of the
flow rate and the geometrical design (Ratio of

Excerpt from the Proceedings of the 2014 COMSOL Conference in Curitiba

large diameter to small diameter, and length of

entrance spacing) on the mass transfer of the
internal wall of the outer tube of reactor. The
study focuses on the mass transfer in the flow
developing and fully developed zones. Table 1
shows the different correlations found in
literature for mass transfer in the annular reactor.

non-reactive with constant physical properties

and that the reactor is operating in steady state
conditions. Also, as the saturation concentration
of the benzoic acid is very low, it is assumed that
it does not change the physical properties of the
water or the flow profile. This allowed solving
the flow independently from the mass transfer:
the CFD model was solved in two steps: First,
the equations of continuity and motion were
solved for getting the flow field across the
computational domain. Then, the velocity values
were kept frozen and the equation of
conservation of species was solved using the
converged flow solution. This solving strategy
saves computation time and brings stability to
the solution.
The equations solved by the Laminar Flow
interface are the Navier-Stokes equations for
conservation of momentum and the continuity
equation for conservation of mass:

2. Governing equation
In this study, the dissolution of a sparingly
soluble wall (benzoic acid) was used to simulate
the mass transfer from the wall. Esteban et al
[31] used successfully this technique in the Finite
Volume analysis with the commercial software
Fluent to determine the mass transfer in the
annular reactor. The laminar flow model was
used to study the flow within the range
200<Re<2100 and the Low Reynolds Number
turbulence model was used for the range
2100<Re<12000.
It is assumed that the fluid is Newtonian fluid
(water), incompressible, isothermal (298 K) and

Table (1): Some correlations reported in literature for mass transfer in annuli.
Hydrodynamic Condition

Ref.

Correlation

Laminar Flow
1

Fully developed

[32]

Shav 1.614 ( Re.Sc.d e /l ) 3

Developing

[17]

Shav 1.029 Re 0.55 .Sc 3 .( d e / l ) 0.472

[33]

Shav 2.703 ( Re.Sc.d e / l ) 3

(i)

(ii)
(iii)

Turbulent Flow
Fully developed

Developing

[32]

Shav 0.023 Re 0.8 .Sc 3

[34]

Shav 0.027 Re 0.8 .Sc 3 .( d 2 / d 1 ) 0.53

[35]

Shav 0.145 Re 2 / 3 .Sc 3 .( d e / l ) 0.25

[17]

Shav 0.095 Re

[33]

Shav 0.305 Re 3 .Sc 3 .( d e / l ) 3

[34]

(iv)

0.85
2

1
3

.Sc .( d e / l ) 0.472
1

3
Shav 0.032 Re .Sc . 1 ( d e / l ) ( d 2 / ( d 1 ) 0.53

0.8

1
3

Excerpt from the Proceedings of the 2014 COMSOL Conference in Curitiba

(v)
(vi)
(vii)
(viii)
(ix)

u j

0
u j

x j
x j
t
u i
u i
p ij
u j

x j
x i x j
t
where is the density, u is the velocity vector, p
is the pressure, and is the viscous stress tensor.
For the case of the turbulent flow, the Reynoldsaveraged Navier-Stokes equations (RANS) were
used.
U j
0
x j

U i

U j

v
U i
u i
P T ij

u j

x j
x i
x j
x j

where U is the averaged velocity field, P is the

averaged pressure, and T is the averaged viscous
stress tensor.
The Low Reynolds Number (LRN k-) modelling
approach has been chosen to solve the turbulent
flow. It has an embedded wall damping effect
that enables the flow modelling to be extended to
the wall without the need to the wall equations
used with the other turbulent flow models (i.e k, k-) [31, 36].
For the mass transfer step, the transport of
diluted species module was used. The
concentration at the dissolving wall was kept
constant at the saturation concentration of the
benzoic acid. The mass transport equation
incorporated the mass transfer due to convection
using the flow profile calculated and to diffusion
-10
using the value of Dm equal to 9.3*10 m2/s
[37]
The conservation of species A assuming constant
density and applying Reynolds method of
averaging of the instantaneous fluctuating
concentration is expressed as:
C A

U j CA
t
x j
x j

C A
u j /CA/
D m

x
j
j

where CA indicates the concentration of the

benzoic acid and the bar indicates a time
averaged value. uj/ and CA/ are the fluctuating
flow velocity and species A concentration
respectively. The u /j C A/

turbulent mass transport, and in analogy with

Ficks law. It is assumed that:

u C J
'

'

t
A

D t C A

The term Dt is called the turbulent diffusivity or

eddy diffusivity. It is not a physical property of
the fluid mixture like Dm but depends on the
character of flow, mainly intensity of turbulence
and on position in the system varying
considerably from the turbulent core to the phase
boundary. The quantity like momentum, energy
and concentration are transported by turbulent
eddies, i.e. the same mechanism, so there should
be a correlation between the parameters
appearing in their flux equations. Applying the
Reynolds analogy between the turbulent
momentum transfer and the turbulent mass
transfer a dimensionless number is defined called
the turbulent Schmidt number:

Sc t

t
D t

In the computational fluid dynamics, many

turbulence models have been developed which
enable the determination of the eddy viscosity.
The Sc(t) dimensionless number typically is close
to unity [38-41] and it was decided to use the
value of 0.9 in the following calculations.

3. Geometrical Model
The annular reactor geometry studied in the
present work is shown in figure (1). The reactor
main dimensions are: 30 mm outer tube
diameter, 20 mm inner tube diameter, and 12
mm inlet diameter tube. The inlet port of the
reactor was centred on the front plate and it was
placed 30 mm away from the inner tube rounded
front. The annular reactor had two sections on
the inner wall of the outer tube that could be set
at constant concentration: The first section
corresponding to the reactor inlet is named
section A, while the other section is named
section B. This feature allowed for studying the
mass transfer process at different hydrodynamic
conditions along the length of the reactor. The
design can be also simplified by using a 2D
axisymmetric model to replace the 3 D model
(figure 2).

terms describe the

Excerpt from the Proceedings of the 2014 COMSOL Conference in Curitiba

Fig. (4): Mesh utilized for the 2D axisymmetric

model.

Fig. (1): Schematic diagram of the 3D annular

reactor model.

The utilized mesh was verified to give mesh

independent results: the CFD simulation results
for the average mass transfer coefficient were
compared with the calculated resulted from the
correlation presented by Mobarak et al. [17] as
shown in table (2).
Table (2): Percentage difference in the average
mass transfer coefficient predicted by Mobarak et
al.[17] and the CFD results for several mesh
refinement for the 2D model.

Fig. (2): Schematic diagram of the 2D

axisymmetric annular reactor model, illustrating
the longitudinal cut plane in the 3D model.

4. Mesh Design
For the system of the mass transfer from the wall
to a turbulent fluid and for high Schmidt number,
the concentration boundary layer is much smaller
than the velocity boundary layer. This makes it
necessary to use a very fine mesh within the near
wall region using the boundary layer mesh
option in Comsol. The 3D geometry is shown in
Figure (3), whereas the 2D axisymmetric mesh is
shown in figure (4).

Mesh

Percent Difference

Extra coarse (4317 elements)

12.97

Coarser (6742 elements)

4.43

Coarse (9331 elements)

1.89

Normal (16631 elements)

1.31

Fine (23610 elements)

1.31

5. Boundary conditions
The boundary conditions for the CFD model
were defined as follows. At the inlet, the mass
flow rate of the fluid was specified and the
direction was defined normal to the boundary.
The hydraulic diameter was fixed at 10 mm and
the turbulence intensity (TI) was set with values
between five and ten percent as recommended in
the Comsol CFD module user guide [41]. At
all the walls, a no slip boundary condition was
imposed. Also, zero diffusive flux of species was
specified at the wall, except for the walls coated
with benzoic acid, where constant concentration
3
of 27.76 mol/m was fixed. This value
corresponds to the saturation concentration of
benzoic acid in water at 298 K [42].
The flow was either solved using the laminar
flow or the low Reynolds Number turbulence
models, then the velocity profile was kept frozen
and the transport of diluted species was used to
solve for the mass transfer from the wall.

Fig. (3): Mesh utilized for the 3D geometry model.

Excerpt from the Proceedings of the 2014 COMSOL Conference in Curitiba

6. Results
6.1. CFD hydrodynamic simulation:
Figure (5) shows the velocity profile along the
annular reactor for (a) laminar flow
corresponding to a flow rate of 1.05 Litres per
minute (Re = 500), (b) turbulent flow
corresponding to a flow rate of 21.04 Litres per
minute (Re = 10000). It has to be mentioned that
Reynolds Number was calculated based on the
equivalent diameter. As seen in figure, the abrupt
expansion and change of direction at the inlet
zone generate large velocity gradient near the
surface. The figure also shows the development
of the velocity profile from the developing flow
throughout the reactor inlet to the developed
flow far away from the inlet effects along the
annular volume.

Fig. (6): Streamlines of velocity field on the

longitudinal center plane at the inlet region using
(a) low Reynolds k turbulence model, (b)
Standard k turbulence model.

6.2. Average Mass Transfer

Performing a mass balance over the reactor to
calculate the rate of mass transfer from the
surface yields the following equation for the
average mass transfer coefficient.
Q C - Ci
K av = ln sat

A C sat - C o
where Q is the flow rate, Ci and Co are the
concentrations of benzoic acid at the inlet and
outlet of the reactor, respectively. The results of
this work showed agreement with those from
other reported investigations where similar
annular configurations were used. Figure (7) and
figure (8) compare the CFD predicted average
mass transfer coefficients with values given by
some of the correlations reported in table (1) for
laminar and turbulent flows respectively.

Fig. (5): Velocity magnitude (m/s) (a) laminar flow

(Re = 500), (b) turbulent flow (Re = 10000).

The velocity profile was also calculated using the

standard k- model of turbulence to compare the
resultant profile with the low Reynolds model.
Both of them showed similar results as seen in
figure (6).

Fig. (7): Comparison of the CFD predicted average

mass transfer coefficients for laminar flow with the
ones estimated using different correlations
reported in the literature (Table (1)).

Excerpt from the Proceedings of the 2014 COMSOL Conference in Curitiba

Fig. (8): Comparison of the CFD predicted average

mass transfer coefficients for transitional and
turbulent flows with the ones estimated using
different correlations reported in the literature
(Table (1)).

It is observed that the CFD predicted average

mass transfer coefficients in section A were
much higher than those in section B. Mass
transfer coefficient in section A was nearly twice
that of section B indicating the importance of the
developing flow on the total mass transfer. These
results suggest that the abrupt expansion and
change of direction that the fluid experiences at
the inlet zone generate high turbulence, as well
as a large near surface concentration gradient,
which results in high mass transfer in section A.
However, as the diffusion layer becomes fully
developed downstream of the entrance in section
B, the mass transfer decreases.
A plot of the CFD predicted value of the average
mass transfer coefficient in the developing flow
region for varying reactor length is given in
figure (9). This figure shows the high value of
the mass transfer coefficient at the leading edge
(corresponding to nearly zero mass transfer
boundary region) and to decrease very rapidly
over the first few centimetres.
The study of the effect of the annulus diameter
ratio (i.e. diameter of outer pipe to diameter of
inner pipe) on the mass transfer coefficient is
shown in figure (10) for section A and figure
(11) for section B. The mass transfer coefficient
can be seen to increase as the annulus diameter
ratio decrease. This can be interpreted as when
the difference between the annulus inlet and
outlet diameters decreases, the surface area to
volume ratio increase, which has a significant
influence on the external mass transfer and
increases the efficiency of the annular reactor.

Fig. (9): Variation of the average mass transfer

coefficient with the change in reactor length.

Fig. (10): Mass transfer coefficient vs. Reynolds

number in section A for three different annulus
diameter ratios.

Fig. (11): Mass transfer coefficient vs. Reynolds

number in section B for three different annulus
diameter ratios.

Excerpt from the Proceedings of the 2014 COMSOL Conference in Curitiba

The effect of the change in position of the inner

tube from the inlet port on the average mass
transfer coefficient in section A is shown in
figure (12).

very small axial distance (x 0.005 m) some

under prediction with respect to the correlations
was found. This may be attributed to the
weakness of the turbulence model in predicting
and capturing all the small eddies generated at
the inlet due to sudden expansion, separation and
reattachment of the flow

Fig. (12): Variation of the average mass transfer

coefficient with the change in inlet spacing.

It is seen that as the inlet spacing decreases the

mass transfer coefficient increases. This result
can be interpreted by the fact that the decrease in
the inlet spacing increases the intensity of abrupt
expansion that the fluid experiences at the inlet
zone, which in turn generates higher turbulence,
as well as a larger near surface concentration
gradient, which results in a higher mass transfer
in section A.

6.3. Local Mass Transfer

The local mass transfer coefficient can be
estimated from the experimental correlation
reported in literature by differentiating the
average mass transfer coefficient by the axial
distance, x [31]:

Kx =

d K av,x
dx

Figure 13 maps the local mass flux magnitude in

section as computed utilizing the laminar flow
model (Re=500).
Figure (14) presents the results obtained for the
reactor operating at Re = 10000. As it can be
seen in the figure, CFD computed local mass
transfer coefficients were consistent with those
calculated from the correlations. However, for

Fig (13): Local mole flux magnitude in section A

for Re=500.

Fig. (14): CFD predictions of the local mass

transfer coefficient along the reactor axis at Re =
10000 compared with values calculated from
correlations.

7. Conclusions
In this study, CFD simulations were carried out
in order to predict surface mass transfer in
annular reactors within a range of flow rates
corresponding to 200 < Re < 12,000 using the
laminar flow model and the Low Reynolds
Number k turbulence model. The work

Excerpt from the Proceedings of the 2014 COMSOL Conference in Curitiba

focused on the geometrical design factors that

will affect the mass transfer in the developing
flow zone and the fully developed flow zone.
The CFD predicted mass transfer data obtained
from this work showed good agreement with
those from other reported investigations where
similar annular configurations were used.
The simulations
following results:

performed

revealed

2.

3.

4.

6.

the

The surface mass transfer taking place near

the entrance region of the reactor is much
higher than that happening in the rest of the
reactor volume. For that reason, this section
should not be ignored when considering the
design of an annular reactor.
The annulus diameter ratio was proved to
have an important impact on the
performance of the annular reactor. It is
proved that when the difference between the
annulus diameters decreases, the rate of
mass transfer increases.
As the inlet spacing between the inlet port of
the reactor and the inner tube rounded front
decreases, the rate of mass transfer
increases.

7.

8.

9.

10.

11.

12.

References
1.

5.

Dalai, A.K., M.N. Esmai, and N.N.

Bakhshi,
Carbon
monoxide
hydrogenation over cobalt catalyst in a
tube-wall reactor: Part II. Modelling
studies. The Canadian Journal of
Chemical Engineering, 1992. 70(2): p.
278-285.
Deutschmann, O. and L.D. Schmidt,
Two-dimensional modeling of partial
oxidation of methane on rhodium in a
short contact time reactor. Symposium
(International) on Combustion, 1998.
27(2): p. 2283-2291.
Kapteijn, F., R.M. de Deugd, and J.A.
Moulijn, FischerTropsch synthesis
using monolithic catalysts. Catalysis
Today, 2005. 105(34): p. 350-356.
Houzelot J., V.J., Mass transfer in
annular cylindrical reactors in laminar

13.

14.

15.

flow. Chemical Engineering Science,

1977. 32: p. 1465 - 1470.
Bryan, P., Advances in understanding
hydrodynamic effects on corrosion.
Corrosion Science, 1993. 35(1-4): p.
655-665.
Allan P, C., A method of correlating
forced convection heat-transfer data
and a comparison with fluid friction.
International Journal of Heat and Mass
Transfer, 1964. 7(12): p. 1359-1384.
Yabuki, A., Near-wall hydrodynamic
effects related to flow-induced localized
corrosion. Materials and Corrosion,
2009. 60(7): p. 501-506.
El Shazly, Y.M.S., Mass transfer
controlled corrosion of baffles in
agitated vessels. Corrosion Engineering,
Science and Technology, 2011. 46(6):
p. 701-705.
Dickinson, C.F. and G.R. Heal, Solid
liquid
diffusion
controlled
rate
equations. Thermochimica Acta, 1999.
340-341(0): p. 89-103.
Saroha, A.K., Solidliquid mass
transfer studies in trickle bed reactors.
Chemical Engineering Research and
Design, 2010. 88(56): p. 744-747.
Askew, W.S. and R.B. Beckmann, Heat
and Mass Transfer in an Agitated
Vessel. Ind. Eng. Chem. Process Des.
Dev., 1965. 4(3): p. 311318.
Shen, G.C., C.J. Geankoplis, and R.s.
Brodkey, A note on particleliquid
mass transfer in a fluidized bed of small
irregular-shaped benzoic acid particles.
Chemical Engineering Science, 1985.
40(9): p. 1797-1802.
Mazhar, H., et al., Experimental
investigation of mass transfer in 90
pipe bends using a dissolvable wall
technique. International Journal of Heat
and Mass Transfer, 2013. 65(0): p. 280288.
Zaki, M.M., I. Nirdosh, and G.H.
Sedahmed,
Mass
transfer
characteristics of reciprocating screen
stack electrochemical reactor in
relation to heavy metal removal from
dilute solutions. Chemical Engineering
Journal, 2007. 126(2-3): p. 67-77.
Berger, F.P. and K.F.F.L. Hau, Mass
transfer in turbulent pipe flow measured

Excerpt from the Proceedings of the 2014 COMSOL Conference in Curitiba

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

by the electrochemical method.

International Journal of Heat and Mass
Transfer, 1977. 20(11): p. 1185-1194.
Sara, O.N., et al., Electrochemical mass
transfer between an impinging jet and a
rotating disk in a confined system.
International Communications in Heat
and Mass Transfer, 2008. 35(3): p. 289298.
Mobarak, A.A., H.A. Farag, and G.
H.Sedahmed, Mass transfer in smooth
and rough annular ducts under
developing flow conditions. Journal of
Applied Electrochemistry, 1997. 27(2):
p. 201-207.
Sonetaka, N., et al., Characterization of
adsorption uptake curves for both
intraparticle diffusion and liquid film
mass transfer controlling systems.
Journal of Hazardous Materials, 2009.
165(1-3): p. 232-239.
Tan, C.S. and J.M. Smith, A dynamic
method for liquid-particle mass transfer
in trickle beds. AIChE Journal, 1982.
28(2): p. 190-195.
Kear, G., et al., Determination of
diffusion controlled reaction rates at a
solid/liquid interface using scanning
electron microscopy. Journal of
Microscopy, 2007. 226(3): p. 218-229.
Yunus A. Cengel, J.M.C., Fluid
Mechanics
Fundamentals
and
Applications 2006, New York: McGraw
- Hill
Date,
A.W.,
Introduction
to
computational fluid dynamics. 2005,
New York: Cambridge university press.
Wang, J. and S.A. Shirazi, A CFD
based correlation for mass transfer
coefficient in elbows. International
Journal of Heat and Mass Transfer,
2001. 44(9): p. 1817-1822.
Davis, C. and P. Frawley, Modelling of
erosioncorrosion
in
practical
geometries. Corrosion Science, 2009.
51(4): p. 769-775.
Khosravi Nikou, M.R. and M.R. Ehsani,
Turbulence models application on CFD
simulation of hydrodynamics, heat and
mass transfer in a structured packing.
International Communications in Heat
and Mass Transfer, 2008. 35(9): p.
1211-1219.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

Moraveji, M.K., et al., Experimental

investigation and CFD simulation of
turbulence effect on hydrodynamic and
mass transfer in a packed bed airlift
internal loop reactor. International
Communications in Heat and Mass
Transfer, 2011. 38(4): p. 518-524.
Ould-Rouis, M., et al., Etude numrique
et exprimentale des transferts de
matire et de quantit de mouvement
dans
un
coulement
annulaire
laminaire non tabli. International
Journal of heat and mass transfer, 1995.
38(6): p. 953-967.
Farias, S.R., Legentilhomme, P., &
Legrand, J. , Finite element simulation
of mass transfer in laminar swirling
decaying flow induced by means of a
tangential inlet in an annulus.
Computer
Methods
in
Applied
Mechanics and Engineering, 2001. 190:
p. 4713 - 4731.
Imoberdorf, G., Cassano, A., Irazoqui,
H., & Alfano, O. , Optimal design and
modeling of annular photocatalytic wall
reactors. Catalysis Today, 2007. 129: p.
118126.
Legrand, J., & Martemyanov, S. A.,
mass transfer in small radius ratio
annuli.
Journal
of
Applied
Electrochemistry, 1994. 24(8): p. 737
744.
Esteban Duran, J., F. Taghipour, and M.
Mohseni, CFD modeling of mass
transfer
in
annular
reactors.
International Journal of Heat and Mass
Transfer, 2009. 52(23-24): p. 53905401.
Ross, T.K. and A.A. Wragg,
Electrochemical mass transfer studies
in annuli. Electrochimica Acta, 1965.
10(11): p. 1093-1106.
Ghosh, U. and S. Upadhyay, Mass
Transfer to Newtonian and NonNewtonian Fluids in Short Annuli.
AIChE Journal, 1985. 31(10): p. 1721 1724.
Rai, B.N., et al., Forced convective
mass transfer in annuli. Chemical
Engineering Communications, 1988.
68(1): p. 15-30.
Pickett, D.J., Electrochemical reactor
design / David J. Pickett. Chemical

Excerpt from the Proceedings of the 2014 COMSOL Conference in Curitiba

10

36.

37.

38.

39.

40.

41.
42.

engineering monographs ; v. 9. 1979:

Elsevier Pub. Co. .
Frei, W.
http://www.comsol.com/blogs/whichturbulence-model-should-choose-cfdapplication/. 2013 [cited 2014 24/05].
Noulty, R.A. and D.G. Leaist, Diffusion
coefficient of aqueous benzoic acid at
25.degree.C. Journal of Chemical &
Engineering Data, 1987. 32(4): p. 418420.
Launder, B.E., Heat and mass
transport, in Turbulence, P. Bradshaw,
Editor.
1978,
Springer
Berlin
Heidelberg. p. 231-287.
Koeltzsch, K., The height dependence
of the turbulent Schmidt number within
the boundary layer. Atmospheric
Environment, 2000. 34(7): p. 11471151.
Reynolds, A.J., The prediction of
turbulent Prandtl and Schmidt numbers.
International Journal of heat and mass
transfer, 1975. 18(9): p. 1055-1069.
Comsol, Comsol 4.3 CFD Module
User's guide. 2012.
Yalkowsky, S.H., Handbook of
Aqueous Solubility Data. 2003: Taylor
& Francis.

Excerpt from the Proceedings of the 2014 COMSOL Conference in Curitiba